非下采样Contourlet变换的彩色图像融合

为了提高图像融合的效果,以多聚焦的彩色图像为研究对象,提出了一种基于非下采样Contourlet变换的多聚焦彩色图像融合算法。对IHS模型的亮度分量I进行NSCT分解,根据多聚焦图像的成像特点和分解后的高低频系数相关性,对分解后的低频系数采用"区域能量取大"和高频系数采用"绝对值和取大"的融合准则进行融合,再进行非下采样Contourlet逆变换,得到融合后的亮度分量,并由此计算融合后的色调分量和饱和度分量,实现彩色图像的融合。实验结果表明,该方法的融合效果优于小波变换的融合结果,针对IHS模型的融合结果也明显优于RGB模型的融合结果。

Color image fusion algorithm based on NSCT

To improve the results of image fusion,a multifocus color image fusion method is presented,which is based on the nonsubsampled Contourlet transform with the particular research on the multi-focus images.Firstly,the intensity component of IHS model is decomposed by NSCT.The decomposed low and high frequency coefficients are fused by the rules of "region energy maximum selectivity" and "absolute value sum maximum selectivity" according to the multi-focus imaging features and the correlation of low frequency coefficients and high frequency coefficient.Then,the fused intensity component can be obtained by the nonsubsampled Contourlet inverse transform to the fused low and high frequency coefficients.Finally,computing the hue component and saturation component of the fused image by the fused intensity component and finishing the whole fused process of the multi-focus color image.The experimental results show that the proposed image fusion method has a better fused quality than that fused by the wavelet transform and the fused results in IHS model is of better quality than that fused in RGB model.